Palaeography or palæography, also spelled paleography (from Greek παλαιός palaiós, “old” and γράφειν graphein, “to write”), is the study of ancient writing. Included in the discipline is the practice of deciphering, reading, and dating historical manuscripts, and the cultural context of writing, including the methods with which writing and books were produced, and the history of scriptoria.
Palaeography can be an essential skill for historians and philologists, as it tackles two main difficulties. First, since the style of a single alphabet in each given language has evolved constantly, it is necessary to know how to decipher its individual characters as they existed in various eras. Second, scribes often used many abbreviations, usually so as to write more quickly and sometimes to save space, so the specialist-palaeographer must know how to interpret them. Knowledge of individual letter-forms, ligatures, punctuation, and abbreviations enables the palaeographer to read and understand the text. The palaeographer must know, first, the language of the text (that is, a 21st-century English or French speaker must become expert in the relevant earlier forms of these languages); and second, the historical usages of various styles of handwriting, common writing customs, and scribal or notarial abbreviations. Philological knowledge of the language, vocabulary, and grammar generally used at a given time or place can help palaeographers identify ancient or more recent forgeries versus authentic documents.
Knowledge of writing materials is also essential to the study of handwriting and to the identification of the periods in which a document or manuscript may have been produced. An important goal may be to assign the text a date and a place of origin: this is why the palaeographer must take into account the style and formation of the manuscript and the handwriting used in it.
Ancient Near East
- Anatolian hieroglyphs
- Cuneiform script
- Egyptian hieroglyphs
- Middle Bronze Age alphabets
- South Arabian alphabet
The Aramaic language was the international trade language of the Ancient Middle East, originating in what is modern-day Syria, between 1000 and 600 BC. It spread from the Mediterranean coast to the borders of India, becoming extremely popular and being adopted by many people, both with or without any previous writing system. The Aramaic script was written in a consonantal form with a direction from right to left. The Aramaic alphabet, a modified form of Phoenician, was the ancestor of the modern Arabic and Hebrew scripts, as well as the Brāhmī script, the parent writing system of most modern abugidas in India, Southeast Asia, Tibet, and Mongolia. Initially, the Aramaic script did not differ from the Phoenician, but then the Aramaeans simplified some of the letters, thickened and rounded their lines: a specific feature of its letters is the distinction between d and r. One innovation in Aramaic is the matres lectionis system to indicate certain vowels. Early Phoenician-derived scripts did not have letters for vowels, and so most texts recorded just consonants. Most likely as a consequence of phonetic changes in North Semitic languages, the Aramaeans reused certain letters in the alphabet to represent long vowels. The letter aleph was employed to write /ā/, he for /ō/, yod for /ī/, and vav for /ū/.
Aramaic writing and language supplanted Babylonian cuneiform and Akkadian language, even in their homeland in Mesopotamia. The wide diffusion of Aramaic letters led to its writing being used not only in monumental inscriptions, but also on papyrus and potsherds. Aramaic papyri have been found in large numbers in Egypt, especially at Elephantine – among them are official and private documents of the Jewish military settlement in 5 BC. In the Aramaic papyri and potsherds, words are separated usually by a small gap, as in modern writing. At the turn of the 3rd to 2nd centuries BC, the heretofore uniform Aramaic letters developed new forms, as a result of dialectal and political fragmentation in several subgroups. The most important of these is the so-called square Hebrew block script, followed by Palmyrene, Nabataean, and the much later Syriac script.
Aramaic is usually divided into three main parts:
- Old Aramaic (in turn subdivided into Ancient, Imperial, Old Eastern and Old Western Aramaic)
- Middle Aramaic, and
- Modern Aramaic of the present day.
The term Middle Aramaic refers to the form of Aramaic which appears in pointed texts and is reached in the 3rd century AD with the loss of short unstressed vowels in open syllables, and continues until the triumph of Arabic.
Old Aramaic appeared in the 11th century BC as the official language of the first Aramaean states. The oldest witnesses to it are inscriptions from northern Syria of the 10th to 8th centuries BC, especially extensive state treaties (c. 750 BC) and royal inscriptions. The early Old Ancient should be classified as “Ancient Aramaic” and consists of two clearly distinguished and standardised written languages, the Early Ancient Aramaic and the Late Ancient Aramaic. Aramaic was influenced at first principally by Akkadian, then from the 5th century BC by Persian and from the 3rd century BC onwards by Greek, as well as by Hebrew, especially in Palestine. As Aramaic evolved into the imperial language of the Neo-Assyrian Empire, the script used to write it underwent a change into something more cursive. The best examples of this script come from documents written on papyrus from Egypt. About 500 BC, Darius I (522–486) made the Aramaic used by the Achaemenid imperial administration into the official language of the western half of the Persian Empire. This so-called “Imperial Aramaic” (the oldest dated example, from Egypt, belonging to 495 BC) is based on an otherwise unknown written form of Ancient Aramaic from Babylonia. In orthography, Imperial Aramaic preserves historical forms – alphabet, orthography, morphology, pronunciation, vocabulary, syntax and style are highly standardised. Only the formularies of the private documents and the Proverbs of Ahiqar have maintained an older tradition of sentence structure and style. Imperial Aramaic immediately replaced Ancient Aramaic as a written language and, with slight modifications, it remained the official, commercial and literary language of the Near East until gradually, beginning with the fall of the Persian Empire (331 BC) and ending in the 4th century AD, it was replaced by Greek, Persian, the eastern and western dialects of Aramaic and Arabic, though not without leaving its traces in the written form of most of these. In its original Achaemenid form, Imperial Aramaic is found in texts of the 5th to 3rd centuries BC. These come mostly from Egypt and especially from the Jewish military colony of Elephantine, which existed at least from 530 to 399 BC.
- See also
A history of Greek handwriting must be incomplete owing to the fragmentary nature of evidence. If one rules out the inscriptions on stone or metal, which belong to the science of epigraphy, we are practically dependent for the period preceding the 4th or 5th century AD on the papyri from Egypt (cf. papyrology), the earliest of which take back our knowledge only to the end of the 4th century BC. This limitation is less serious than might appear, since the few manuscripts not of Egyptian origin which have survived from this period, like the parchments from Avroman or Dura, the Herculaneum papyri, and a few documents found in Egypt but written elsewhere, reveal a uniformity of style in the various portions of the Greek world; but some differences can be discerned, and it is probable that, were there more material, distinct local styles could be traced.
Further, any given period several types of hand may exist together. There was a marked difference between the hand used for literary works (generally called “uncials” but, in the papyrus period, better styled “book-hand”) and that of documents (“cursive“) and within each of these classes several distinct styles were employed side by side; and the various types are not equally well represented in the surviving papyri.
The development of any hand is largely influenced by the materials used. To this general rule the Greek script is no exception. Whatever may have been the period at which the use of papyrus or leather as a writing material began in Greece (and papyrus was employed in the 5th century BC), it is highly probable that for some time after the introduction of the alphabet the characters were incised with a sharp tool on stones or metal far oftener than they were written with a pen. In cutting a hard surface, it is easier to form angles than curves; in writing the reverse is the case; hence the development of writing was from angular letters (“capitals”) inherited from epigraphic style to rounded ones (“uncials”). But only certain letters were affected by this development, in particular E (uncial ε), Σ (c), Ω (ω), and to a lesser extent A (α).
The earliest Greek papyrus yet discovered is probably that containing the Persae of Timotheus, which dates from the second half of the 4th century BC and its script has a curiously archaic appearance. E, Σ, and Ω have the capital form, and apart from these test letters the general effect is one of stiffness and angularity. More striking is the hand of the earliest dated papyrus, a contract of 311 BC. Written with more ease and elegance, it shows little trace of any development towards a truly cursive style; the letters are not linked, and though the uncial c is used throughout, E and Ω have the capital forms. A similar impression is made by the few other papyri, chiefly literary, dating from about 300 BC; E may be slightly rounded, Ω approach the uncial form, and the angular Σ occurs as a letter only in the Timotheus papyrus, though it survived longer as a numeral (= 200), but the hands hardly suggest that for at least a century and a half the art of writing on papyrus had been well established. Yet before the middle of the 3rd century BC, one finds both a practised book-hand and a developed and often remarkably handsome cursive.
These facts may be due to accident, the few early papyri happening to represent an archaic style which had survived along with a more advanced one; but it is likely that there was a rapid development at this period, due partly to the opening of Egypt, with its supplies of papyri, and still more to the establishment of the great Alexandrian Library, which systematically copied literary and scientific works, and to the multifarious activities of Hellenistic bureaucracy. From here onward, the two types of script were sufficiently distinct (though each influenced the other) to require separate treatment. Some literary papyri, like the roll containing Aristotle‘s Constitution of Athens, were written in cursive hands, and, conversely, the book-hand was occasionally used for documents. Since the scribe did not date literary rolls, such papyri are useful in tracing the development of the book-hand.
The documents of the mid-3rd century BC show a great variety of cursive hands. There are none from chancelleries of the Hellenistic monarchs, but some letters, notably those of Apollonius, the finance minister of Ptolemy II, to this agent, Zeno, and those of the Palestianian sheikh, Toubias, are in a type of script which cannot be very unlike the Chancery hand of the time, and show the Ptolemaic cursive at its best. These hands have a noble spaciousness and strength, and though the individual letters are by no means uniform in size there is a real unity of style, the general impression being one of breadth and uprightness. H, with the cross-stroke high, Π, Μ, with the middle stroke reduced to a very shallow curve, sometimes approaching a horizontal line, Υ, and Τ, with its cross-bar extending much further to the left than to the right of the up-stroke, Γ and Ν, whose last stroke is prolonged upwards above the line, often curving backwards, are all broad; ε, c, θ and β, which sometimes takes the form of two almost perpendicular strokes joined only at the top, are usually small; ω is rather flat, its second loop reduced to a practically straight line. Partly by the broad flat tops of the larger letters, partly by the insertion of a stroke connecting those (like H, Υ) which are not naturally adapted to linking, the scribes produced the effect of a horizontal line along the top of the writing, from which the letters seem to hang. This feature is indeed a general characteristic of the more formal Ptolemaic script, but it is specially marked in the 3rd century BC.
The Derveni Papyrus, a Greek Macedonian philosophical text dating around 340 BC, considered Europe’s oldest manuscript
Besides these hand of Chancery type, there are numerous less elaborate examples of cursive, varying according to the writer’s skill and degree of education, and many of them strikingly easy and handsome.[according to whom?] In some cursiveness is carried very far, the linking of letters reaching the point of illegibility, and the characters sloping to the right. A is reduced to a mere acute angle (∠), T has the cross-stroke only on the left, ω becomes an almost straight line, H acquires a shape somewhat like h, and the last stroke of N is extended far upwards and at times flattened out until it is little more than a diagonal stroke to the right. The attempt to secure a horizontal line along the top is here abandoned. This style was not due to inexpertness, but to the desire for speed, being used especially in accounts and drafts, and was generally the work of practised writers. How well established the cursive hand had now become is shown in some wax tablets of this period, the writing on which, despite the difference of material, closely resemble the hands of papyri.
Documents of the late 3rd and early 2nd centuries BC show, perhaps partly by the accident of survival (there is nothing analogous to the Apollonius letters, a loss of breadth and spaciousness. In the more formal types the letters stand rather stiffly upright, often without the linking strokes, and are more uniform in size; in the more cursive they are apt to be packed closely together. These features are more marked in the hands of the 2nd century. The less cursive often show am approximation to the book-hand, the letters growing rounder and less angular than in the 3rd century; in the more cursive linking was carried further, both by the insertion of coupling strokes and by the writing of several letters continuously without raising the pen, so that before the end of the century an almost current hand was evolved. A characteristic letter, which survived into the early Roman period, is T, with its cross-stroke made in two portions (variants:). In the 1st century, the hand tended, so far as can be inferred from surviving examples, to disintegrate; one can recognise the signs which portend a change of style, irregularity, want of direction, and the loss of the feeling for style. A fortunate accident has preserved two Greek parchments written in Parthia, one dated 88 BC, in a practically unligatured hand, the other, 22/21 BC, in a very cursive script of Ptolemaic type; and though each has non-Egyptian features the general character indicates a uniformity of style in the Hellenistic world.
The development of the Ptolemaic book-hand is difficult to trace, as there are few examples, mostly not datable on external grounds. Only for the 3rd century BC have we a secure basis. The hands of that period have an angular appearance; there is little uniformity in the size of individual letters, and though sometimes, notably in the Petrie papyrus containing the Phaedo of Plato, a style of considerable delicacy is attained, the book-hand in general shows less mastery than the contemporary cursive. In the 2nd century the letters grew rounder and more uniform in size, but in the 1st century there is perceptible, here as in the cursive hand, a certain disintegration. Probably at no time did the Ptolemaic book-hand acquire such unity of stylistic effect as the cursive.
Papyri of the Roman period are far more numerous and show greater variety. The cursive of the 1st century has a rather broken appearance, part of one character being often made separately from the rest and linked to the next letter. A form characteristic of the 1st and 2nd century and surviving after that only as a fraction sign (=⅛) is η in the shape . By the end of the 1st century, there had been developed several excellent types of cursive, which, though differing considerably both in the forms of individual letters and in general appearance, bear a family likeness to one another. Qualities which are specially noticeable are roundness in the shape of letters, continuity of formation, the pen being carried on from character to character, and regularity, the letters not differing strikingly in size and projecting strokes above or below the line being avoided. Sometimes, especially in tax-receipts and in stereotyped formulae, cursiveness is carried to an extreme. In a letter of the prefect, dated in 209, we have a fine example of the Chancery hand, with tall and laterally compressed letters, ο very narrow and α and ω often written high in the line. This style, from at least the latter part of the 2nd century, exercised considerable influence on the local hands, many of which show the same characteristics less pronounced; and its effects may be traced into the early part of the 4th century. Hands of the 3rd century uninfluenced by it show a falling off from the perfection of the 2nd century; stylistic uncertainty and a growing coarseness of execution mark a period of decline and transition.
Several different types of book-hand were used in the Roman period. Particularly handsome[according to whom?] is a round, upright hand seen, for example, in a British Museum papyrus containing Odyssey III. The cross-stroke of ε is high, Μ deeply curved and Α has the form α. Uniformity of size is well attained, and a few strokes project, and these but slightly, above or below the line. Another type, well called by palaeographer Schubart the “severe” style, has a more angular appearance and not infrequently slopes to the right; though handsome, it has not the sumptuous appearance of the former. There are various classes of a less pretentious style, in which convenience rather than beauty was the first consideration and no pains were taken to avoid irregularities in the shape and alignment of the letters. Lastly may be mentioned a hand which is of great interest as being the ancestor of the type called (from its later occurrence in vellum codices of the Bible) the biblical hand. This, which can be traced back at least the late 2nd century, has a square, rather heavy appearance; the letters, of uniform size, stand upright, and thick and thin strokes are well distinguished. In the 3rd century the book-hand, like the cursive, appears to have deteriorated in regularity and stylistic accomplishment.
In the charred rolls found at Herculaneum and dating from about the beginning of our era, are specimens of Greek literary hands from outside Egypt; and a comparison with the Egyptian papyri reveals great similarity in style and shows that conclusions drawn from the henads of Egypt may, with caution, be applied to the development of writing in the Greek world generally.
The cursive hand of the 4th century shows some uncertainty of character. Side by side with the style founded on the Chancery hand, regular in formation and with tall and narrow letters, which characterised the period of Diocletian, and lasted well into the century, we find many other types mostly marked by a certain looseness and irregularity. A general progress towards a florid and sprawling hand is easily recognisable, but a consistent and deliberate style was hardly evolved before the 5th century, from which unfortunately few dated documents have survived. Byzantine cursive tends to an exuberant hand, in which the long strokes are excessively extended and individual letters often much enlarged. But not a few hands of the 5th and 6th centuries are truly handsome and show considerable technical accomplishment. Both an upright and a sloping type occur and there are many less ornamental hands, but there gradually emerged towards the 7th century two general types, one (especially used in letters and contracts) a current hand, sloping to the right, with long strokes in such characters at τ, ρ, ξ, η (which has the h shape), ι, and κ, and with much linking of letters, and another (frequent in accounts), which shows, at least in essence, most of the forms of the later minuscule. (cf. below.) This is often upright, though a slope to the right is quite common, and sometimes, especially in one or two documents of the early Arab period, it has an almost calligraphic effect.
In the Byzantine period, the book-hand, which in earlier times had more than once approximated to the contemporary cursive, diverged widely from it.
Vellum and paper manuscripts
The change from papyrus to vellum involved no such modification in the forms of letters as followed that from metal to papyrus. The justification for considering the two materials separately is that after the general adoption of vellum, the Egyptian evidence is first supplemented and later superseded by that of manuscripts from elsewhere, and that during this period the hand most used was one not previously employed for literary purposes.
The prevailing type of book-hand during what in papyrology is called the Byzantine period, that is, roughly from AD 300 to 650, is known as the biblical hand. It went back to at least the end of the 2nd century and had had originally no special connection with Christian literature. In manuscripts, whether vellum or paper, of the 4th century found in Egypt are met other forms of script, particularly a sloping, rather inelegant hand derived from the literary hand of the 3rd century, which persisted to at least the 5th century; but the three great early codices of the Bible are all written in uncials of the biblical type. In the Vaticanus, placed in the 4th century, the characteristics of the hand are least strongly marked; the letters have the forms characteristic of the type but without the heavy appearance of later manuscripts, and the general impression is one of greater roundness. In the Sinaiticus, which is not much later, the letters are larger and more heavily made; and in the Alexandrinus (5th century) a later development is seen, with emphatic distinction of thick and thin strokes. By the 6th century, alike in vellum and in papyrus manuscripts, the heaviness had become very marked, though the hand still retained, in its best examples, a handsome appearance; but after this it steadily deteriorated, becoming ever more mechanical and artificial. The thick strokes grew heavier; the cross strokes of T and Θ and the base of Δ were furnished with drooping spurs. The hand, which is often singularly ugly, passed through various modifications, now sloping, now upright, though it is not certain that these variations were really successive rather than concurrent. A different type of uncials, derived from the Chancery hand and seen in two papyrus examples of the Festal letters despatched annually by the Patriarch of Alexandria, was occasionally used, the best known example being the Codex Marchalianus (6th or 7th century). A combination of this hand with the other type is also known.
The uncial hand lingered on, mainly for liturgical manuscripts, where a large and easily legible script was serviceable, as late as the 12th century, but in ordinary used it had long been superseded by a new type of hand, the minuscule, which originated in the 8th century, as an adaptation to literary purposes of the second of the types of Byzantine cursive mentioned above. A first attempt at a calligraphic use of this hand, seen in one or two manuscripts of the 8th or early 9th century, in which it slopes to the right and has a narrow, angular appearance, did not find favour, but by the end of the 9th century a more ornamental type, from which modern Greek script descended, was already established. It has been suggested that it was evolved in the Monastery of Stoudios at Constantinople. In its earliest examples it is upright and exact but lacks flexibility; accents are small, breathings square in formation, and in general only such ligatures are used as involve no change in the shape of letters. The single forms have a general resemblance (with considerable differences in detail) both to the minuscule cursive of late papyri, and to those used in modern Greek type; uncial forms were avoided.
In the course of the 10th century the hand, without losing its beauty and exactness, gained in freedom. Its finest period was from the 9th to the 12th century,[according to whom?] after which it rapidly declined. The development was marked by a tendency
- to the intrusion, in growing quantity, of uncial forms which good scribes could fit into the line without disturbing the unity of style but which, in less expert hands, had a disintegrating effect;
- to the disproportionate enlargement of single letters, especially at the beginnings and ends of lines;
- to ligatures, often very fantastic, which quite changed the forms of letters;
- to the enlargement of accents, breathings at the same time acquiring the modern rounded form.
But from the first there were several styles, varying from the formal, regular hands characteristic of service books to the informal style, marked by numerous abbreviations, used in manuscripts intended only for a scholar’s private use. The more formal hands were exceedingly conservative, and there are few classes of script more difficult to date than the Greek minuscule of this class. In the 10th, 11th and 12th centuries a sloping hand, less dignified than the upright, formal type, but often very handsome, was especially used for manuscripts of the classics.
Hands of the 11th century are marked in general (though there are exceptions) by a certain grace and delicacy, exact but easy; those of the 12th by a broad, bold sweep and an increasing freedom, which readily admits uncial forms, ligatures and enlarged letters but has not lost the sense of style and decorative effect. In the 13th and still more in the 14th centuries there was a steady decline; the less formal hands lost their beauty and exactness, becoming ever more disorderly and chaotic in their effect, while formal style imitated the precision of an earlier period without attaining its freedom and naturalness, and often appears singularly lifeless. In the 15th century, especially in the West, where Greek scribes were in request to produce manuscripts of the classical authors, there was a revival, and several manuscripts of this period, though markedly inferior to those of the 11th and 12th centuries, are by no means without beauty.
Accents, punctuation, and division of words
In the book-hand of early papyri, neither accents nor breathings were employed. Their use was established by the beginning of the Roman period, but was sporadic in papyri, where they were used as an aid to understanding, and therefore more frequently in poetry than prose, and in lyrical oftener than in other verse. In the cursive of papyri they are practically unknown, as are marks of punctuation. Punctuation was effected in early papyri, literary and documentary, by spaces, reinforced in the book-hand by the paragraphos, a horizontal stroke under the beginning of the line. The coronis, a more elaborate form of this, marked the beginning of lyrics or the principal sections of a longer work. Punctuation marks, the comma, the high, low and middle points, were established in the book-hand by the Roman period; in early Ptolemaic papyri, a double point (:) is found.
In vellum and paper manuscripts, punctuation marks and accents were regularly used from at least the 8th century, though with some differences from modern practice. At no period down to the invention of printing did Greek scribes consistently separate words. The book-hand of papyri aimed at an unbroken succession of letters, except for distinction of sections; in cursive hands, especially where abbreviations were numerous, some tendency to separate words may be recognised, but in reality it was phrases or groups of letters rather than words which were divided. In the later minuscule word-division is much commoner but never became systematic, accents and breathings serving of themselves to indicate the proper division.
Uncial is a majuscule script (written entirely in capital letters) commonly used from the 4th to 8th centuries AD by Latin and Greek scribes. Uncial letters were used to write Greek, Latin, and Gothic.
Early uncial script is likely to have developed from late Old Roman cursive. Early forms are characterized by broad single stroke letters using simple round forms taking advantage of the new parchment and vellum surfaces, as opposed to the angular, multiple stroke letters, which are more suited for rougher surfaces, such as papyrus. In the oldest examples of uncial, such as the De bellis macedonicis manuscript in the British Library, all of the letters are disconnected from one another, and word separation is typically not used. Word separation, however, is characteristic of later uncial usage.
As the script evolved over the centuries, the characters became more complex. Specifically, around AD 600, flourishes and exaggerations of the basic strokes began to appear in more manuscripts. Ascenders and descenders were the first major alterations, followed by twists of the tool in the basic stroke and overlapping. By the time the more compact minuscule scripts arose circa AD 800, some of the evolved uncial styles formed the basis for these simplified, smaller scripts. Uncial was still used, particularly for copies of the Bible, tapering off until around the 10th century. There are over 500 surviving copies of uncial script, by far the largest number prior to the Carolingian Renaissance.
In general, there are some common features of uncial script:
- m, n, and u are relatively broad; m is formed with curved strokes (although a straight first stroke may indicate an early script), and n is written as N to distinguish it from r and s.
- f, i, p, s, t are relatively narrow.
- e is formed with a curved stroke, and its arm (or hasta) does not connect with the top curve; the height of the arm can also indicate the age of the script (written in a high position, the script is probably early, while an arm written closer to the middle of the curve may indicate a later script).
- l has a small base, not extending to the right to connect with the next letter.
- r has a long, curved shoulder, often connecting with the next letter.
- s resembles (and is the ancestor of) the “long s“; in uncial it looks more like r than f.
In later uncial scripts, the letters are sometimes drawn haphazardly; for example, double-l runs together at the baseline, bows (for example in b, p, r) do not entirely curve in to touch their stems, and the script is generally not written as cleanly as previously.